Add liscence and prepare for OSS release
🚦 CI / changes (push) Successful in 39s
🚦 CI / docker (push) Has been skipped
🚦 CI / test (push) Has been skipped
🚦 CI / docs (push) Successful in 7s

This commit is contained in:
2026-06-28 12:05:56 +02:00
parent 7c6a8be2b7
commit a4de64ea04
5 changed files with 93 additions and 12 deletions
+33 -3
View File
@@ -289,7 +289,7 @@ not data.
Overhead µs/item at **work_us = 10** (framework overhead dominates):
| Topology | KPN private | TBB |
| Topology | KPN++ | TBB |
|---|---|---|
| chain depth-1 | 1.7 | **1.4** |
| chain depth-4 | 2.5 | **2.2** |
@@ -301,7 +301,7 @@ Overhead µs/item at **work_us = 10** (framework overhead dominates):
Overhead µs/item at **work_us = 100** (moderate compute, KPN wins):
| Topology | KPN private | TBB |
| Topology | KPN++ | TBB |
|---|---|---|
| chain depth-1 | **2.1** | 3.5 |
| chain depth-4 | **4.3** | 5.2 |
@@ -311,7 +311,7 @@ Overhead µs/item at **work_us = 100** (moderate compute, KPN wins):
| wide fanout-4 | 3.4 | **1.9** |
| diamond (2×2) | **4.1** | 6.1 |
KPN private pools beat TBB for every chain and diamond topology at 100 µs/node, and
KPN++ pools beat TBB for every chain and diamond topology at 100 µs/node, and
match TBB within ~20% at 10 µs/node for shallow chains. TBB retains an edge on wide
fanout (serial dispatch loop vs. work-stealing pool) and at extreme oversubscription
depths (chain-32 at 10 µs). The remaining gap at light work is the cost of
@@ -400,3 +400,33 @@ examples/
scripts/
render_readme.py — regenerates README.md from README.md.in
```
---
## Contributing
Contributions are welcome. This project is hosted on a self-hosted Gitea
instance that accepts sign-in and registration with a GitHub account, so you
can log in with your existing GitHub identity to open issues and pull requests.
If you change any code that appears in a README snippet, edit `README.md.in`
(the template) rather than `README.md` directly, then regenerate:
```bash
cmake --build build --target readme # or: python scripts/render_readme.py
```
---
## Acknowledgments
AI tooling was used heavily throughout the development of this project,
including the design, implementation, tests, and documentation. All output
has been reviewed, but please keep this in mind when reading or building on the
code.
---
## License
Released under the [MIT License](LICENSE). Copyright (c) 2026 Duncan Tourolle.